Job SummaryMoultrie is looking for a Data Scientist to join the growing Data and Analytics Team. This team owns the development of insights from extraction that power decision-making across Moultrie. This role will build the predictive and prescriptive modeling capabilities that sit on top of the foundational data layer to surface actionable insights and recommendations within BI products and downstream systems. You will own the statistical modeling and feature engineering that exists in the Data
- Data Layer & Feature Store: Snowflake
- Model Deployment: Snowflake ML Functions, Snowpark
- Development: Python, SQL
- Feature store: Snowflake
- Experiment Tracking: MLflow
Typical work includes building and maintaining feature stores Snowflake/dbt, training and validating predictive models against curated datasets, and delivering model outputs as attributes made available in datasets ready for downstream tools.
The ideal candidate has hands-on experience with applied predictive modeling in a business context (churn prediction, customer health scores, propensity scoring) and can contribute to the data work required to support it. This role will work closely with the other members of the Data and Analytics team as well as business stakeholders to align work with highest business impact.
Job ResponsibilitiesPredictive and Prescriptive Modeling - Design, build, and maintain predictive models that address defined business questions (customer churn, subscription health, purchase propensity) using data from the foundational layer in Snowflake.
- Deliver model outputs as attributes that can integrate cleanly into downstream BI products (Tableau, Streamlit) and activation platforms (BlueConic, Braze, TripleWhale).
- Work with business stakeholders to translate ambiguous questions into scoped modeling problems with defined success metrics.
- Communicate model outputs and their business implications clearly to non-technical audiences.
- Track, document, and monitor experiments and deployed models to ensure outputs are reliable, understandable, and reproduceable.
Feature Store and Data Engineering - Contribute to the build and maintenance of features in the feature store: defining features, documenting refresh cadence, and ensuring feature pipelines are reliable and tested.
- Build and maintain training datasets with clear documentation of assumptions, evaluation windows, and limitations.
- Work with data engineers to ensure data models are structured to support feature engineering and model training.
- Ability to apply data engineering fundamentals (SQL modelling, versional control, documentation) to contribute to the feature store and build statistical models that integrate into the existing foundational tech stack.
Job RequirementsSkills and Qualifications - 4+ years of hands-on experience in data science or a role with significant applied modelling.
- Demonstrated experience building and deploying predictive models in a business context.
- Strong Python proficiency: Proven experience using libraries (scikit-learn, pandas) to build and maintain predictive models.
- Experience with classification and regression techniques and the ability to validate model performance using appropriate metrics (precision, recall, AUC, etc.).
- SQL proficiency: Able to write clean, maintainable code for supporting models in Snowflake with support from data engineers.
- Experience with Git and standard software development practices: version control, code reviews, branching, and CI/CD basics.
- Ability to take an ambiguous business problem and work backwards to produce models that support effective solutions by creating a list of requirements and working through sprints to deliver.
- Strong documentation practices: able to produce and maintain model definitions, lineage documentation, and data dictionaries that enable other developers and business stakeholders.
- Collaborative working style: comfortable operating at the boundary between data engineering, data science, and business teams.
Preferred Qualifications - Experience in retail, CPG, or consumer hardware.
- Hands-on experience with a feature store platform.
- Experience with MLflow or a comparable experiment tracking tool.
- Familiarity with Snowpark or Snowflake ML Functions.
- Experience delivery model outputs that can be used in BI tools and other downstream platforms.
